[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20181129183550.GD30281@lst.de>
Date: Thu, 29 Nov 2018 19:35:50 +0100
From: Christoph Hellwig <hch@....de>
To: Brian Starkey <Brian.Starkey@....com>
Cc: Christoph Hellwig <hch@....de>, "Clark, Rob" <robdclark@...il.com>,
Dave Airlie <airlied@...ux.ie>,
linux-arm-msm <linux-arm-msm@...r.kernel.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
dri-devel <dri-devel@...ts.freedesktop.org>,
Tomasz Figa <tfiga@...omium.org>,
Sean Paul <seanpaul@...omium.org>,
"vivek.gautam@...eaurora.org" <vivek.gautam@...eaurora.org>,
freedreno <freedreno@...ts.freedesktop.org>,
Robin Murphy <Robin.Murphy@....com>, nd <nd@....com>
Subject: Re: [PATCH v3 1/1] drm: msm: Replace dma_map_sg with dma_sync_sg*
On Thu, Nov 29, 2018 at 05:33:03PM +0000, Brian Starkey wrote:
> This sounds very useful for ion, to avoid CPU cache maintenance as
> long as the buffer stays in device-land.
>
> One question though: How would you determine "the last user to unmap"
> to know when to do the final "make visible to CPU" step?
I'd assume the user of the DMA API keeps track of it. E.g. dmabuf
would be a pretty natural layer to implement common logic for this.
Powered by blists - more mailing lists